Oestrogen (E2) is known to promote the proliferation of thyroid papillary carcinoma cells (KAT5). However, the molecular mechanism responsible is not well understood. In the study reported herein, the localization of ER alpha (ERalpha) and beta (ERbeta) in KAT5 and anaplastic carcinoma cells (FRO) was studied by immunofluorescence staining and by immunoblotting the proteins in subcellular fractions. Cell proliferation and apoptosis were also determined together with the expression of relevant proteins. The pattern of the subcellular localization of ERalpha and ERbeta differed between papillary and anaplastic cancer. Upon E2 treatment, the level of ERalpha increased in the nuclei of papillary cancer cells but ERbeta remained unchanged. The level of mitochondrial ERbeta surpassed that of ERalpha in anaplastic cancer cells. The different locations of ERalpha and ERbeta in KAT5 and FRO agreed with the finding that E2 promoted the proliferation of KAT5 but inhibited or did not affect that of FRO cells, and with the proposed functions of these two receptors. E2 inhibited the level of Bax in the mitochondria of papillary cancer, followed by a decrease of cytochrome c and/or apoptosis-inducing factor (AIF) release from the mitochondria into the cytosol. However, in anaplastic cancer, E2 promoted the expression of Bax in the mitochondria and the release of cytochrome c and/or AIF from mitochondria into the cytosol. Our results may explain the differences in epidemiology and responses to anti-tumour therapy between papillary and anaplastic cancer in terms of the subcellular localization of ER isoforms. In conclusion, the findings provide evidence to support the observation that E2 is an important factor in the development of thyroid cancer. The subcellular localization of ERalpha and ERbeta may account for the different pathogenesis of thyroid papillary and anaplastic cancers.

5-Hydroxymethylcytosine (5hmC) is a recently characterized epigenetic mark that is particularly abundant in brain tissue and that regulates gene transcription. We have recently begun to understand the important role of 5hmC in brain development, plasticity and disease, but there are currently little data on 5hmC alterations in psychiatric illnesses. Here we report what we believe to be the first genome-wide analysis of 5hmC in the depressed brain. Using AbaSI sequencing, we investigated 5hmC in the prefrontal cortex of depressed (N=19) and psychiatrically healthy controls (N=19). Consistent with previous global 5hmC analyses in other phenotypes, and likely owing to the inter-individual variability in 5hmC content, the distribution of 5hmC across chromosomes and genomic features was not different between groups. We did, however, find 550 CpGs with suggestive evidence of differential hydroxymethylation. Of these, we validated CpGs in the gene body of myosin XVI (MYO16) and insulin-degrading enzyme using targeted oxidative bisulfite sequencing. Furthermore, the enrichment of 5hmC was also associated with changes in the expression of these two genes in depressed suicides. Together, our results present a novel mechanism linking increased 5hmC to depression and provide a framework for future research in this field.

Conversion of cholesterol to biologically active steroids is a multi-step enzymatic process. Along with some important enzymes, like cholesterol side-chain cleavage enzyme (P450scc) and 3 -hydroxysteroid dehydrogenase/ isomerase (3 -HSD), several proteins play key role in steroidogenesis. The role of steroidogenic acute regulatory (StAR) protein is well established. A novel protein, BRE, found mainly in brain, adrenals and gonads, was highly expressed in hyperplastic rat adrenals with impaired steroidogenesis, suggesting its regulation by pituitary hormones. To further elucidate its role in steroidogenic tissues, mouse Leydig tumor cells (mLTC-1) were transfected with BRE antisense probes. Morphologically the BRE antisense cells exhibited large cytoplasmic lipid droplets and failed to shrink in response to human chorionic gonadotropin. Although cAMP production, along with StAR and P450scc mRNA expression, was unaffected in BRE antisense clones, progesterone and testosterone yields were significantly decreased, while pregnenolone was increased in response to human chorionic gonadotropin stimulation or in the presence of 22(R)OH-cholesterol. Furthermore, whereas exogenous progesterone was readily converted to testosterone, pregnenolone was not, suggesting impairment of pregnenolone-to-progesterone conversion, a step metabolized by 3 -HSD. That steroidogenesis was compromised at the 3 -HSD step was further confirmed by the reduced expression of 3 -HSD type I (3ß-HSDI) mRNA in BRE antisense cells compared with controls. Our results suggest that BRE influences steroidogenesis through its effects on 3 -HSD action, probably affecting its transcription.